Battelle Northwest Laboratories

The aqueous solubility of Fe₂(SeO₃)₃·6H₂O(c) was studied in deionized water adjusted to a range in pH values from 0.77 to 5.1 and in Na₂SeO₃ solutions ranging in concentrations from 0.0002 to 0.02 mol-dm^-3.The studies were conducted from both the undersaturation and oversaturation directions, with equilibration periods ranging from 7 to 1725 days.Stoichiometric dissolution of the solid was observed in solutions with pH values up to nearly 4.In general, concentrations of both Se and Fe decreased as pH increased from 1 to 4.Analyses of the equilibrated suspensions confirmed solid to be Fe₂(SeO₃)₃·6H₂O(c) and the aqueous Se to be selenite.Pitzer's ion-interaction model was used with selected ion pairs to interpret the solubility data.The logarithm of the solubility product of ferric selenite Fe₂(SeO₃)₃·6H₂O(c) ⇄ 2Fe³⁺ + 3SeO₃^2- + 6H₂O was found to be -41.58 ± 0.11.This value is less than any reported in the literature for a ferric selenite by more than 10 orders of magnitude.The solubility data and calculations show an extremely strong interaction between aqueous Fe³⁺ and SeO₃^2-; interpretation of these data requires the inclusion of FeSeO₃⁺ i.e. Fe³⁺ + SeO₃^2- ⇄ FeSeO₃⁺, log K = 11.15 ± 0.11.

Soil samples containing coal tar residues from manufactured gas (town gas) plants were extracted by carbon dioxide supercritical fluid extraction (SFE) and Soxhlet extraction techniques.The detection limits and the reproducibility of the SFE method were evaluated, and the comparability of measured concentrations of polycyclic aromatic hydrocarbons (PAH) by the two extraction methods was determinedLaboratory extracts were analyzed by high-resolution gas chromatog. to quantify selected PAH.SFE recoveries of soil-spiked PAH compounds were quant. within exptl. error.The detection limits of the method were in the 50-100 ppb range.The reproducibility in PAH concentrations determined for three early SFE replicate extracts averaged ±25% standard deviation of the mean.A later reproducibility study for 10 replicate SFE extracts of a different sample averaged ±16% standard deviation of the mean (12% due to anal.; 4% due to SFE replication).Five Soxhlet extracts averaged ±11% standard deviation from the mean reproducibility (10% due to anal.; 1% due to replication).SFE extract PAH concentrations varied from Soxhlet extract PAH concentrations by an average ±20%, ±12%, and ±10% for three samples.Greater than 80% and 100% extraction efficiencies were achieved in the first 30 min of SFE when three successive extractions were performed on two samples, resp.SFE tended to discriminate against high mol. weight PAH due to their lower solubilities in supercritical carbon dioxide.Overall, the SFE method provided rapid (30-min) extraction that may prove to be useful for in-the-field determinations of organic compounds in soils.The SFE method shows promise as an alternative to, or an addition to, traditional Soxhlet extraction